From: Waleed Oransa (WORANSA@eg.ibm.com)
Date: Thu Jan 10 2008 - 11:09:49 CST
"Philippe Verdy" <verdy_p@wanadoo.fr> wrote on 10/01/2008 04:09:55 Õ:
> Behnam [mailto:behnam.rassi@gmail.com] wrote:
> > An rtl paragraph can not be anything else, wherever it goes. This is
> > what I have such a hard time to make understand to non rtl users. It
> > isn't about page layout design. It's about preserving the integrity
> > of the text itself, as it was written by the writer.
>
> And I do think that an RTL paragraph can stay RTL independently of the
page
> layout, even if it happens to be left-aligned, because the same would
also
> happen to a Latin-text presented in a centered page layout, or justified
> layout, or right-aligned layout, without changing the text itself.
>
> There's absolutely no difference with what would happen to a Latin text
> within a general Arabic page layout for example. The solutions to keep
the
> integrity of the text are the same in both cases.
Please note that pure Arabic or Hebrew text is not the common case. The
common
case today which cause a lot of pain to Bidi users is the mixed Bidi and
Latin
text.
>
> This is exactly the same issue with the presentation layout of numbers
in
> tables: the numbers do not need separate encoding just because they
appear
> to be right-aligned or aligned on the decimal separator (where a part of
the
> number is right-aligned and the rest is left-aligned: both parts are the
> same number, encoded identically, because the alignment is not essential
to
> their actual meaning or reading).
The alignment problem is minor compared to the text direction.
> My opinion is then that the problems you are experimenting do not come
from
> Unicode itself, but from the tools or libraries or computer languages
you
> use to represent to compute the page layout: such problems or
deficiencies
> must be resolved in those tools or libraries or languages.
>
> These problems seem to be solved in HTML, and in wellknown GUI
construction
> libraries or document formats, for example Java/Swing, X11/Motif, RTF,
> MSOffice/Word, OpenOffice.
As you mentioned the problem is solved by different ways, I would say that
many created
an encoding for their different languages but Unicode provided the best
way to make
all people communicate with each other in one Universal encoding. Because
no standard way
required to store the direction, each one invent his own way to solve the
problem.
This archive was generated by hypermail 2.1.5 : Thu Jan 10 2008 - 11:12:19 CST